1. Technical Field of the Invention
The present invention relates to an apparatus and method for processing a speech signal, specifically to an apparatus and method for processing a speech signal which include what is called a function for converting a speech rate in order to help the elderly to easily listen to a speech.
The present invention also relates to a method and apparatus for communicating speeches, specifically to a method and apparatus which converts speech rates in a speech communicating system, such as a cellular phone system, which do not always communicate speeches in clear receiving conditions.
2. Background Art
It is a proved fact that generally the elderly tend to have difficulties in comprehension of speech uttered rapidly in comparison with their youth. As aids for the elderly, an apparatus and method for processing a speech signal have been known which include what is called a function of converting a speech rate. The function of converting a speech rate is pursued as follows. A pause in a voiced speech is identified out of an inputted speech signal. By use of a time brought by the pause, the speech which has been made in the course of phonation is extended in a temporal axis without modulating the pitch height instead of shortening the paused time. Accordingly, the voiced speech is converted into a slower speech as a whole. (See Japanese Patent Application Laid-open No.Hei.8-146985, for example).
In an apparatus for processing a speech signal according to the Japanese Patent Application Laid-open No.Hei.8-146985, speech rate controlling information for controlling a speech rate is stored in data which is going to be transmitted, sound recording media and the like in advance. Then, a speech rate is controlled on a basis of the speech rate controlling information in a sound reproducing apparatus for receiving and reproducing the transmitted data or a sound reproducing apparatus for the sound recording media. In addition, a radio which includes the function of converting a speech rate has been developed (see Imai, Takagi, Yomogida, Takeishi. “Choshukinou wo Sonaeta Rajio no Kaihatu [Development of a radio which includes a function of hearing aids].” (The Institute of Electronics, Information and Communication Engineers), IEICE trans.TL2003-7, Jun. 2003).
Incidentally, this kind of technologies of a speech rate conversion has been energetically studied by broadcasting agencies, too. It is conceivable that transmitting persons transmit speech signals which have been obtained by subjecting speeches to a speech rate conversion processes for the convenience of the listening elderly in the field of speech signals transmitted in television and radio broadcastings in the future. With regard to aired programs for which transmitting persons have converted a speech rate, therefore, it will be likely that a speech rate conversion is duplicated by a function of converting a speech rate which is implemented by the recipients.
This causes problems that a speech signal which has been obtained by subjecting speech to a speech rate conversion process is additionally treated with another speech rate conversion process, and accordingly that the converted speech rate is slower than is necessary so that the speech unexpectedly becomes more difficult to listen to. With regard to a conventional receiver, these problems are solved if users turn off the function of converting a speech rate manually. For each aired program, however, users have to turn on and off the function of converting a speech rate depending on whether or not the programs has been subjected to a speech rate conversion process. This will likely causes the users to feel annoyed. In addition, it is not practical that the elderly are expected to turn on and off the function of converting a speech rate while receivers are in use.
On the other hand, a telephone set to which technologies of converting a speech rate are applied has been proposed and translated into practical use. The telephone set converts a speech rate of a counterpart of an incoming call (see Japanese Patent Application Laid-open No.Hei.2001-268175, for example).
A conventional apparatus for communicating speeches which is installed into the telephone set according to the invention disclosed in Japanese Patent Application Laid-open No.Hei.2001-268175 automatically identifies a counterpart by use of a service of displaying counterparts' telephone numbers. By this, a temporal axis of a speech signal of the counterpart who has been identified is extended corresponding to a speech rate which has been registered for each counterpart in advance.
The conventional apparatus for communicating speeches which is installed into the telephone set is intended to cause a telephone set to subject a speech signal of a counterpart to a speech rate conversion process. In a case that this conventional apparatus is installed into a cellular phone, however, it is likely that a speech signal from the counterpart includes much noise or is partially interrupted depending on radio wave conditions (i.e. conditions for receiving calls). If such a speech signal which has been received under such bad receiving conditions is subjected to a speech rate conversion process, however, this will likely bring about a problem that the speech unexpectedly becomes more difficult to listen to.